This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright Author's personal copy

Biomaterials 32 (2011) 1787e1795

Contents lists available at ScienceDirect

Biomaterials

journal homepage: www.elsevier.com/locate/biomaterials

Activation of stress-regulated transcription factors by triethylene glycol dimethacrylate monomer

Stephanie Krifka a, Christine Petzel a, Carola Bolay a, Karl-Anton Hiller a, Gianrico Spagnuolo b, Gottfried Schmalz a, Helmut Schweikl a,* a Department of Operative Dentistry and Periodontology, University of Regensburg, D-93042 Regensburg, Germany b Department of Oral and Maxillofacial Sciences, University of Naples “Federico II”, Italy article info abstract

Article history: Triethylene glycol dimethacrylate (TEGDMA) is a resin monomer available for short exposure scenarios of Received 22 October 2010 oral tissues due to incomplete polymerization processes of dental composite materials. The generation of Accepted 14 November 2010 reactive oxygen species (ROS) in the presence of resin monomers is discussed as a common mechanism Available online 10 December 2010 underlying cellular reactions as diverse as disturbed responses of the innate immune system, inhibition of dentin mineralization processes, genotoxicity and a delayed cell cycle. Yet, the signaling pathway Keywords: through a network of proteins that finally initiates the execution of monomer-induced specific cell Resin monomer responses is unknown so far. The aim of the present study was to extend the knowledge of molecular TEGDMA Stress response mechanisms of monomer-induced cell death as a basis for reasonable therapy strategies. Thus, the Transcription factor monomer-induced expression and phosphorylation of stress-related transcription factors was analyzed in various cell lines. The time-related induction of apoptosis was investigated as well. The expression of p53 increased in HeLa cell cultures treated with camptothecin (positive control) for 24h, and the formation of p53Ser15 and p53Ser46 was detected in cell nuclei by Western blotting. TEGDMA (3 mM) appeared to stimulate p53 expression only slightly, but increased p21 expression was found in cell nuclei and cytoplasm. Both camptothecin and TEGDMA increased p53 expression to some extent in the nuclear fraction in human transformed pulp-derived cells (tHPC), and similar effects were detected in RAW264.7 macrophages. No clear induction of c-Jun and phospho-c-Jun by TEGDMA was detected in HeLa cell nuclei, and the expression of ATF-2 and phospho-ATF-2 was inhibited in the presence of the monomer. ATF-3 expression was found only in the nuclear fraction of camptothecin-treated HeLa cultures. TEGDMA seemed to inhibit the formation of phospho-c-Jun and phospho-ATF-2 in tHPC, and the monomer acted negatively on the expression of c-Jun, ATF-2 and ATF-3 in RAW264.7 macrophages. These changes in the expression and activation of stress-related transcription factors were time-related to the induction of apoptosis by TEGDMA in all cell lines. The present study provides experimental evidence that TEGDMA interferes with the regulation of cellular pathways through transcription factors activated as a conse- quence of DNA damage like p53 or initiated downstream of MAPK (mitogen-activated protein kinases) like c-Jun, ATF-2 and ATF-3. The direct causal correlation between DNA damage, activation or inhibition of MAPKs and transcription factors, and apoptosis is under current investigation. However, the induction of apoptosis in different cell lines in the presence of monomers like TEGDMA may be subject to a higher level of complexity than currently suggested by simple linear models. Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction composite materials are not chemically inert in the human oral cavity. Residual unreacted monomers are available for short expo- Tissues of the oral cavity are impacted by a variety of biological sure scenarios due to incomplete polymerization processes of the stressors including compounds released from dental restorative organic matrix of dental composites. Triethylene glycol dimetha- biomaterials. Numerous investigations have shown that dental crylate (TEGDMA) was found to be a major comonomer eluted even from polymerized resin composites in an aqueous environment [1]. In addition, polymerized composites are also susceptible to * Corresponding author. Fax: þ49 941 944 6025. biodegradation through hydrolytic enzymes in saliva, making the E-mail address: [email protected] (H. Schweikl).

0142-9612/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2010.11.031 Author's personal copy

1788 S. Krifka et al. / Biomaterials 32 (2011) 1787e1795 polymeric network most likely a source of continuous release of Besides the suggested role of MAPKs in the regulation of cell compounds to which cells and tissues are then exposed for a long death and survival, the function of the tumor suppressor p53 as period of time [2]. a key regulator of apoptosis is firmly established [18]. P53 is acti- Among others, the resin monomers TEGDMA and HEMA caused vated through the ATM signaling pathway when DNA damage specific stress responses in a wide variety of eukaryotic cells in vitro occurs and, in addition to many other functions, acting as a tran- [3]. These substances induced cell death via apoptosis in various cell scription factor controls cell cycle progression and apoptosis-related types including pulp and gingiva cells and it appears as if cell death gene expression or interacts with proteins after posttranslational was mediated, at least in part, by the generation of reactive oxygen modifications including phosphorylation at Ser15 and Ser46 resi- species (ROS) disturbing the redox balance [4,5]. Furthermore, dues. Among the p53 target genes, p21Cip1 expression is activated, genotoxic and mutagenic effects caused by TEGDMA are probably to which in turn acts as an inhibitor of cell proliferation or apoptosis some extent a consequence of ROS-induced DNA damage. A signif- depending on its cellular localization [19]. icant increase in the levels of 8-oxoguanine as a marker for oxidative Although not experimentally proven so far, it has been proposed DNA damage was found in cell cultures after long exposure to that an increase in ROS caused by resin monomers is the main origin TEGDMA [6]. As a result of DNA damage, mammalian cells activate of the upstream activation of MAPK [17]. In contrast, the related functional cell cycle checkpoints through the coordinated activities downstream activation of MAPK-related gene expression of specific of regulatory proteins. A corresponding monomer-induced delay of transcription factors that finally initiate the execution of monomer- the cell cycle was, however, overridden in the presence of the induced apoptosis is unknown to date. The activation of cascades of antioxidant N-acetylcysteine (NAC) [7,8]. Furthermore, low phosphorylation, in many cases increases expression of immediate TEGDMA concentrations and chemically related substances like early genes as an early stress response. These immediate early genes PMMA (poly-methyl methacrylate) even inhibited specific odonto- code for transcription factors, which then regulate downstream blast functions including alkaline phosphatase activity, the matrix genes of the cell survival network [20]. Therefore, it was the aim of mineralizing capability, calcium deposition, and gene expression, the present study to broaden our understanding of the molecular such as dentin sialoprotein [9,10]. Oxidative stress is most likely the mechanisms of dental resin monomer-induced cell death. To this cause of disturbed mineralization because NAC dramatically end, we analyzed the monomer-induced expression and phos- reduced cytotoxicity of bone substitutes to maintain osteoblastic phorylation of transcription factors that were associated with the viability and function, implying enhanced bone regeneration induction of apoptosis. around NAC-treated inorganic biomaterials [11]. The monomer First, we hypothesized that the induction of apoptosis may TEGDMA also influenced specific cell responses of the innate depend on a differential activation of various transcription factors immune system. The monomer instantaneously downregulated the which are targets of the MAPK signaling pathway. For instance, the LPS-induced cytokine production in macrophages and inhibited the activity of the transcription factor AP-1 is rapidly elevated due to expression of surface antigens like CD14 and others surface markers the induction of fos and jun gene transcription by many stimuli essential for the controlled interaction of immune cells [12]. including genotoxic agents [21,22]. In addition to the activation at These findings clearly indicate that the adverse effects of resin the transcriptional level, the activity of c-Jun is greatly enhanced in monomers involve interference with a complex regulatory network response to phosphorylation by the MAPK JNK [23]. Activating of specific cell functions. However, except for the well-established transcription factor 2 (ATF-2), as a member of the ATF/CREB family effects regarding the induction of oxidative stress by resin monomers, of transcription factors, is activated via phosphorylation by MAPKs there is only one additional minor insight available in the research in the cellular response to stress [24]. It has been reported that thus far into the specific mechanisms behind these phenomena, and activation of ATF-2 by p38/JNK is necessary for the induction of the inhibition of specific functions in cells of various origins. apoptosis, and ATF-2 has also been shown to coordinate the cellular A more detailed understanding will provide a better estimation response to DNA damage upon activation of ATM [25,26]. ATF-3 has of the consequences associated with dental therapies. Thus, the been discussed as a stress-inducible gene or an ‘adaptive response’ analysis of the mechanisms of the cytotoxicity of dental composite gene. Its biological role is currently not well defined but probably materials is highly attractive because it leads the way to the related to the regulating of cell proliferation and apoptosis [27]. development of useful strategies and materials for the protection of Second, because of the demonstrated genotoxicity of dental resin tissues of the oral cavity [13]. monomers, monomer-induced apoptosis may also be initiated by Research related to this goal has primarily focused on signal the cellular response to DNA damage followed by the activation of transduction pathways through the mitogen-activated protein a signaling cascade through ATM, p53, and p21 [18]. kinase (MAPK) cascade. These proteins regulate basic cell functions HeLa were included in the current investigation as an established modified by resin monomers like cell viability and apoptosis, cell human cancer cell line and a control for the expression of the p53 proliferation and the expression of cytokines and cell surface anti- tumor suppressor protein. In addition, cell responses were analyzed gens [14]. It has been reported that ERK1/2 was activated by HEMA in cells important for clinical applications and tissue-specific reac- and TEGDMA in a salivary gland cell line after long exposure periods, tions. RAW264.7 mouse macrophages were used as a model cell line and a role of p38 in the apoptotic cell response to HEMA has been of the innate immune system, and human pulp-derived cells were suggested [15]. Moreover, the sustained activation of both ERK1/2 utilized as a model of pulp tissue responses [17]. The dental resin and p38 kinases by TEGDMA in a human monocyte cell line (THP-1) TEGDMA served as a model monomer to study the expression and was associated with monomer-induced cytotoxicity, and a pro- posttranslational modification of transcription factors because of survival role of ERK1/2 was indicated in primary human pulp cells the demonstrated relationship between monomer structure and after short exposure periods to HEMA [6,16]. ERK1/2, p38 and JNK cytotoxicity [3,28]. (c-Jun N-terminal kinase) were differentially activated by phos- phorylation in the presence of lipopolysaccharide (LPS), a known 2. Materials and Methods inducer of MAPK activity, and the resin monomer TEGDMA in RAW264.7 mouse macrophages. In contrast to the immediate 2.1. Chemicals and reagents inhibition of cytokine release, apoptosis and necrosis caused by LPS Triethylene glycol dimethacrylate (TEGDMA; CAS-No. 109-16-0) and and TEGDMA was a late response associated with a strong increase Nonidet P-40 substitute (NP-40; 74385) were purchased from SigmaeAldrich in MAPK activation [17]. (Taufkirchen, Germany), camptothecin was obtained from BioVision (Mountain Author's personal copy

S. Krifka et al. / Biomaterials 32 (2011) 1787e1795 1789

View, CA, USA), and Accutase came from PAA Laboratories GmbH (Cölbe, 1mM NaVO4, protease inhibitor cocktail) incubated on ice, and then centrifuged at Germany). Dulbecco’smodified Eagle’s medium (DMEM) was obtained from 16,000 g for 10 min in the cold. The supernatant was collected as the nuclear Gibco Life Technologies (Karlsruhe, Germany). RPMI 1640 medium containing L- fraction. The amount of protein present in each cytoplasmic and nuclear fraction was glutamine and 2.0 g/l NaHCO3 was purchased from PAN Biotech (Aidenbach, determined by a BCA protein assay (Sigma, Taufkirchen, Germany) using bovine Germany). Minimum essential medium a (MEMa), fetal bovine serum (FBS), serum albumin as a standard. penicillin/streptomycin, and phosphate-buffered saline supplemented with 5 mM EDTA (PBS-EDTA) came from Life Technologies, Gibco BRL (Eggenstein, 2.5. Western blot analysis Germany). Antiphospho-p53Ser15 (no. 9284), antiphospho-p53Ser46 (no. 2521), anti-p53 (1C12), antiphospho-ATF-2 (Thr69/71) (no. 9225), anti-lamin A/C (no. For Western blot analysis, proteins of the cytoplasmic and nuclear fraction 2032), and anti-rabbit IgG HRP-linked antibody (no. 7074) were obtained from (20 mg per lane) were first separated on a 10e12% sodium dodecyl sulfate-poly- Cell Signaling (NEB Frankfurt, Germany). Anti-p21 (sc-756), anti-c-Jun (sc-1694), acrylamide gel by electorophoresis (SDS-PAGE), and transferred to a nitrocellulose anti-ATF-2 (C-19) (sc-187), and anti-ATF-3 (C-19) (sc-187) came from Santa Cruz membrane in SDS-electroblot buffer (25 mM Tris-Cl, 192 mM glycin, 20% methanol, Biotech (Heidelberg, Germany). An alternative anti-p21 antibody (EP1125Y) was pH 8.3) at 350 mA for 60 min. The membrane was then washed twice in TBS (25 mM obtained from Epitomics (Biomol GmbH, Hamburg, Germany), anti-glyceralde- Tris-Cl, 150 mM NaCl, pH 7.4). After blocking with 5% nonfat milk in TBST (TBS plus hyde-3-phosphate dehydrogenase (GAPDH) monoclonal antibody (clone 6C5) 0.1% Tween 20, pH 7.4) at room temperature for 60 min, the membrane was incu- was purchased from Millipore GmbH (Schwalbach, Germany), and goat anti- bated with primary antibodies in 5% nonfat milk in TBST (TBS plus 0.1% Tween 20, pH mouse IgG (H þ L)-HRP conjugate came from Bio-Rad Laboratories (Munich, 7.4) over night at cold temperatures. Then the membrane was washed with TBST Germany). A protease inhibitor cocktail (complete mini) was obtained from three times at room temperature for 10 min. Primary antibodies were detected by Roche Diagnostics (Mannheim, Germany), and a FACS Annexin V-FITC apoptosis appropriate horseradish peroxidase-conjugated secondary antibodies in TBST at detection kit came from R&D Systems (Minneapolis, MN, USA). All other chem- room temperature for 60 min. After washing for 20 min in TBST and 10 min in PBS, icals used in the present study were at least chemical grade. bound secondary antibodies were visualized by enhanced chemiluminescence (ECL). For reprobing, the membranes were first stripped at room temperature for 2.2. Cell culture and exposure of cells 15 min using a CHEMICON re-blot plus mild antibody stripping solution (Millipore, Schwalbach, Germany). RAW264.7 mouse macrophages (ATCC TIB71) were cultivated in RPMI 1640 medium containing Leglutamine, sodium-pyruvate and 2.0 g/l NaHCO3 supple- mented with 10% fetal bovine serum (FBS) and penicillin-streptomycin. Human 3. Results pulp-derived cells (tHPC) were routinely kept in MEMa supplemented with 10% FBS, penicillin (100U/ml), geneticin (50 mg/ml) and streptomycin (100 mg/ml) at 37 C 3.1. Induction of apoptosis in HeLa, human pulp cells and mouse and 5% CO2 as described [29]. HeLa (human cervix adenocarcinoma) cells were macrophages grown in Dulbecco’s modified Eagle’s medium (DMEM) containing 5% FBS, 4.5 g/l glucose, penicillin (100U/ml), and streptomycin (100 mg/ml). The induction of apoptosis by the resin monomer TEGDMA was 2.3. Determination of apoptosis determined in various cell lines to analyze for responses of cells with tissue-specific functions. In HeLa cells, a significant decrease e 5 The cells (0.5 1.0 10 /well) were cultivated in six-well plates for 24 h at in the number of viable cells to about 82% caused by camptothecin 37 C and 5% CO . Then, cell cultures were treated with cell culture medium 2 was related to an increase in the percentage of cells in early containing TEGDMA (1 and 3 mM)or1mM camptothecin for 24 h. The exposure of the cell cultures was stopped by discarding the exposure media, and then washing apoptosis (14%) after a 24 h exposure period (Fig. 1A). Likewise, the the cells with PBS (phosphate-buffered saline) at room temperature. Next, the cell resin monomer TEGDMA induced a concentration-related increase cultures were harvested with PBS-EDTA (HeLa, macrophages) or Accutase (tHPC), in the number of cells in early apoptosis to 3% (1 mM TEGDMA) and centrifuged, washed in PBS, and finally collected by centrifugation. The exposure 8% (3 mM TEGDMA) compared to 1% found in untreated cell of phosphatidylserine on the cell surface as a marker of apoptosis was determined by the binding of annexin V-FITC in 100 ml binding buffer as recommended by the cultures. The percentage of cells in late apoptosis increased to 16% manufacturer. In addition, cells were stained with propidium iodide (PI) to in cell cultures treated with 3 mM TEGDMA, but the proportion of differentiate between cells in apoptosis (annexin V) and necrosis (PI) [17,30]. necrotic cells induced by camptothecin and TEGDMA was below 1% Finally, the stained cell cultures were analyzed on a FACSCanto flow cytometer (Fig. 1A). (Becton Dickinson, San Jose, CA, USA). FITC fluorescence (FL-1) was collected The amount of viable human transformed pulp cells (tHPC) was through a 530/30 band pass filter, and PI fluorescence (Fl-3) through a 650 nm long pass filter. Data acquisition (2 104 events for each sample) was performed reduced to 55% and 38% by 1 and 3 mM TEGDMA, respectively with FACSDivaÔ 5.0.2 software. The same software was used to count the compared to 85% in untreated controls. The proportion of cells in numbers of viable (annexin V-; PI-) cells in the lower left quadrant (Q3) of density late apoptosis increased in parallel from 19% to 36% in cultures plots, and the percentages of cells in apoptosis (annexin Vþ; PI-; lower right þ þ treated with 1 and 3 mM TEGDMA compared to 6% in control quadrant Q4), late apoptosis (annexin V ;PI ; upper right quadrant Q2), and fi necrosis (annexin V-; PIþ; upper left quadrant Q1) were determined accordingly cultures, and a signi cant increase in cells in necrosis was detected [17,30]. in TEGDMA-treated cultures. Camptothecin, however, reduced the Original data from at least four independent experiments were summarized as number of viable tHPC only slightly from 85% to 73%, and medians (25e75% quartiles), and differences between median values were statisti- a significant increase in the number of cells in the various stages of cally analyzed using the ManneWhitney U test (SPSS 18.0, SPSS, Chicago, IL, USA) for cell death was not detected compared to untreated cultures pairwise comparisons among groups at the 0.05 level of significance. (Fig. 1B). 2.4. Analysis of the expression of transcription factors and protein extraction RAW264.7 mouse macrophages were more susceptible to camptothecin than HeLa cells since the number of viable cells was 6 The cells (1.0e2.0 10 ) were seeded into cell culture plates (15 cm in diameter) drastically reduced to 14% compared to 82% in untreated control and left untreated for 24 h at 37 C and 5% CO . Then the medium was replaced by 2 cultures (Fig. 1C). The proportion of cells treated with camptothecin fresh cell culture medium containing TEGDMA (1 and 3 mM)or1mM camptothecin for 24 h at 37 C and 5% CO2. Exposure was stopped by collecting the exposure media found in late apoptosis and necrosis increased to about 61% and 19% and floating cells. Cells which adhered to the cell culture plates were washed with in parallel, and only a relatively small fraction were identified as ice-cold PBS, treated with PBS-EDTA (HeLa, mouse macrophages) or Accutase cells in early apoptosis. TEGDMA caused a concentration-depen- fl fi (tHPC), combined with the oating cells, and nally collected by centrifugation dent high increase in the percentage of cells in apoptosis and together. Next, nuclear and cytosolic extracts were separated by the use of different lysis buffers. First, the cell pellet was resuspended in 1 ml PBS, then 0.5 ml buffer A necrosis. In cell cultures exposed to 1 and 3 mM TEGDMA, 1% and 5% (10 mM Tris HCl, 60 mM KCl, 1 mM Na2EDTA, 1 mM DTT, pH 7.4) was added. The cells of the cells were found in early apoptosis, while 16% and 47% of the were incubated on ice for 5 min and then collected by centrifugation in the cold. The cells were detected in late apoptosis. Even the amount of necrotic supernatant was discarded, the cell pellet was resuspended in buffer B (buffer A plus cells induced by TEGDMA was significantly increased to 19% 0.4% NP 40, 5 mM NaF, 1 mM NaVO , protease inhibitor cocktail), incubated on ice for 4 (Fig. 1C). The percentage of viable cells in TEGDMA-exposed 3 min, and centrifuged for 4 min. The supernatant was collected as a cytoplasmic cell fraction. The pellet washed in buffer A, resuspended in buffer C (20 mM Tris-Cl pH cultures decreased to 82% (1 mM) and 28% (3 mM) compared to 93% 8.0, 400 mM NaCl, 1.5 mM MgCl2, 1.5 mM Na2EDTA, 25% glycerol, 1 mM DTT, 5 mM NaF, found in untreated control cultures. Author's personal copy

1790 S. Krifka et al. / Biomaterials 32 (2011) 1787e1795

3.2. Expression of transcription factors associated with apoptosis

3.2.1. Expression of transcription factors in HeLa A The expression of transcription factors associated with the HeLa Medium 100 induction of apoptosis in the presence of TEGDMA was analyzed in a TEGDMA 1mM the same cell lines. The tumor suppressor protein p53 was not

)%(re c TEGDMA 3mM b detected in either the cell nuclei or cytoplasmic fraction of 80 Camptothecin 1µM untreated HeLa cultures. However, p53 expression was greatly

b enhanced in cultures treated with camptothecin, which was used 60 muNlleC as a positive control after a 24 h exposure period. A very strong single band was detected with a specific antibody in the nuclear 40 fraction (Fig. 2A). In contrast, TEGDMA appeared to increase p53 expression only slightly in cell nuclei at a high concentration b 20 (Fig. 2A). The formation of p53 Ser15 and Ser46 phosphorylation b c was also clearly detectable in nuclei isolated from cell cultures a c b 0 exposed to camptothecin. However, this posttranslational modifi- ni ined nex + PI PI cation of p53 was not unequivocally detected in TEGDMA-treated nsta An exin u Ann HeLa cells under the current experimental conditions. The p53- induced cyclin-dependent kinase (CDK) inhibitor p21 was found in B the cytoplasmic as well as in the nuclear fraction of untreated HeLa cells. Moreover, the expression of p21 was slightly enhanced in 100 tHPC nuclei of cells treated with camptothecin, and TEGDMA appeared to increase p21 expression in both cell nuclei and cytoplasm (Fig. 2A). )%(rebmuNlleC c 80 Lamin A/C and GAPDH were used as markers for the nuclear and the cytoplasmic fraction of cell extracts. Large amounts of lamin A/C a 60 b were detected in the nuclear fraction of untreated and treated cell b 40 a a b 20 a

0 Medium 1 µM1 Camptothecin mM3 TEGDMA mM TEGDMAMedium1 µM1 Camptothecin mM 3TEGDMA mM TEGDMA nied exin + PI PI nsta Ann exin Cytosol Nuclei u Ann A C p53 100 RAW 264.7 Macrophages phospho-p53(Ser15) a phospho-p53(Ser46) )%( 80 p21

r Lamin A/C ebmuNlleC c 60 b 40 b cleaved Lamin A b GAPDH c 20 c a c b B a a 0 c-Jun n nied nexi + PI IP phospho-c-Jun nsta An exin u Ann ATF-2 phospho-ATF-2 Fig. 1. Time-dependent induction of apoptosis and necrosis in HeLa, human pulp- derived cells (tHPC), and RAW264.7 mouse macrophages. Medium ¼ untreated cell ATF-3 cultures. Cell cultures were exposed to 1 and 3 mM TEGDMA, and 1 mM camptothecin for 24 h and analyzed by flow cytometry after staining with annexin V-FITC and Fig. 2. Expression of transcription factors in HeLa. Cell cultures were exposed to 1 and propidium iodide (PI). Bars represent the percentages of viable cells (unstained), cells m in apoptosis (annexin), cells in late apoptosis (annexin þ PI), and cells in necrosis (PI) 3mM TEGDMA and 1 M camptothecin for 24 h, and proteins in cell nuclei and the as medians (25% and 75% percentiles) calculated from five (n ¼ 5) independent cytoplasmic cell fraction were detected as described in Materials and Methods. ¼ ¼ experiments. a ¼ significant differences between median values obtained in untreated Medium untreated cell cultures. (A) transcription factors related to cell responses after DNA damage; (B) ¼ transcription factors related to cell responses through controls (medium) and cell cultures exposed to 1 mM TEGDMA; b ¼ significant differences between median values obtained in untreated controls (medium) and cell mitogen-activated protein kinases (MAPK). Phosphorylated proteins were detected by immunoblotting using specific antibodies, and the expression of the total amount of cultures exposed to 3 mM TEGDMA; c ¼ significant differences between median values obtained in untreated controls (medium) and cell cultures exposed to camptothecin. proteins (phosphorylated and non-phosphorylated) was also monitored on the same membranes after stripping and reprobing with specific antibodies. The expression of Lamin A/C and GAPDH was used as a marker of cell nuclei and the cytoplasmic cell fraction. Typical results from one of at least two independent experiments are shown. Author's personal copy

S. Krifka et al. / Biomaterials 32 (2011) 1787e1795 1791 cultures as expected, while GAPDH was present only in the cyto- plasmic fraction. Noteworthy is that a faint band was present in the nuclear fraction of camptothecin-treated HeLa cell representing the cleavage product of lamin A (Fig. 2A). Transcription factors downstream of the MAPK cascade were differentially expressed after exposure to the resin monomer and Medium1 µM CamptothecinMedium 1 mM3 TEGDMA mM TEGDMA1 µM1 Camptothecin mM3 TEGDMA mM TEGDMA camptothecin. Exposure of HeLa to camptothecin increased the Cytosol Nuclei expression of c-Jun in cell nuclei compared to untreated controls, and phosphorylation of c-Jun was detected as well. Only a very faint A band indicating c-Jun was visible in the cytoplasmic fraction p53 (Fig. 2B). Most importantly, a clear induction of phospho-c-Jun and c-Jun by TEGDMA was not detected in HeLa cells despite a faint phospho-p53(Ser15) band in extracts of cells exposed to 3 mM TEGDMA under the phospho-p53(Ser46) current experimental conditions. In contrast to this pattern, phos- p21 phorylated ATF-2 and total ATF-2 were clearly detectable in the nuclear fraction of cell extracts of untreated cultures. However, the Lamin A/C expression of phospho-ATF-2 and ATF-2 was not activated but appeared to be inhibited by camptothecin and TEGDMA. None of GAPDH the protein species were detectable after the same exposure periods in the cytoplasmic fractions. ATF-3 expression was found only in the nuclear fraction of camptothecin-treated HeLa cell B cultures (Fig. 2B). c-Jun phospho-c-Jun 3.2.2. Expression of transcription factors in tHPC Transformed human pulp-derived cells (tHPC) expressed the ATF-2 p53 protein in both cell nuclei and the cytoplasmic fraction of phospho-ATF-2 untreated cells (Fig. 3A). Camptothecin increased p53 expression to ATF-3 some extent in the nuclear fraction, and it appeared as if 1 mM TEGDMA had a similar effect. Moreover, phosphorylation of p53 at Ser15 was detected in all cell fractions, and these levels increased in Fig. 3. Expression of transcription factors in human pulp-derived cells (tHPC). Cell cultures were exposed to 1 and 3 mM TEGDMA and 1 mM camptothecin for 24 h, and the presence of camptothecin or 1 mM TEGDMA in the nuclear proteins in cell nuclei and the cytoplasmic cell fraction were detected as described in fraction. A weak induction of the phosphorylation of p53 at Ser 46 Materials and Methods. Medium ¼ untreated cell cultures. (A) ¼ transcription factors was indicated in nuclear fractions of cultures treated with camp- related to cell responses after DNA damage; (B) ¼ transcription factors related to cell tothecin and TEGDMA. The p21 protein was hardly detectable in the responses through mitogen-activated protein kinases (MAPK). Phosphorylated proteins were detected by immunoblotting using specific antibodies, and the expres- nuclei of untreated cell cultures, but increased in cultures treated sion of the total amount of proteins (phosphorylated and non-phosphorylated) was with camptothecin and TEGDMA (Fig. 3A). However, an increase in also monitored on the same membranes after stripping and reprobing with specific p21 expression in the cytoplasm caused by TEGDMA was not antibodies. The expression of Lamin A/C and GAPDH was used as a marker of cell nuclei obvious. Lamin A/C was detected in the nuclear fraction of and the cytoplasmic cell fraction. Typical results from one of at least two independent untreated and treated cell cultures as expected, although the band experiments are shown. detected in a cell culture exposed to 3 mM TEGDMA was weaker compared to untreated controls, while GAPDH was detected in TEGDMA concentration (Fig. 4A). Lamin A/C and GAPDH were equal amounts on all cytoplasmic fractions (Fig. 3A). detected in nuclear and cytoplasmic fractions of all cell cultures. A The expression of c-Jun was observed in cell nuclei and the cleavage product of lamin A was present in the nuclear fraction of cytoplasm as well. Camptothecin increased the amounts of phos- RAW264.7 cells exposed to 3 mM TEGDMA (Fig. 4A). phorylated c-Jun in cell nuclei while TEGDMA appeared to inhibit Similarly to the observations found with HeLa and tHPC, the formation of phospho-c-Jun (Fig. 3B). A very similar pattern of camptothecin and TEGDMA acted negatively on the expression of expression was detected with ATF-2 and phospho ATF-2. As with transcription factors downstream from MAPK as suggested in HeLa, expression of ATF-3 was detected only in the nuclear fraction a linear signaling model. The expression of c-Jun was observed in of camptothecin-treated tHPC cultures (Fig. 3B). cell nuclei of untreated controls and a weak expression was detected in the cytoplasmic fraction as well (Fig. 4B). However, c- 3.2.3. Expression of transcription factors in RAW264.7 mouse Jun expression was lower in cell cultures exposed to camptothecin macrophage cells and TEGDMA. The amounts of phospho-c-Jun were reduced in the In RAW264.7 macrophages, p53 was detectable in cell nuclei of nuclear fraction, and no phospho-c-Jun was detected in the cyto- untreated cell cultures (Fig. 4A). Moreover, camptothecin and 1 mM plasmic fraction except for a very faint band in extracts of untreated TEGDMA caused an increase in the expression of p53, and a single cells. The expression of ATF-2 and phosphorylated ATF-2 was band was even detected in the cytoplasmic fraction of campto- inhibited by camptothecin and a high TEGDMA concentration in thecin-treated cells. Furthermore, p53 was phosphorylated at Ser15 cell nuclei, and neither ATF-2 nor phospho-ATF-2 were detected in in the nuclear cell fraction after exposure of cells to camptothecin the cytosol. ATF-3 expression was also inhibited after treatment of and TEGDMA. Again, a distinct single band indicating p53Ser15 was cells with camptothecin and TEGDMA (Fig. 4B). present in the cytosol of cells treated with camptothecin. Unfor- The inhibitory effects of camptothecin and TEGDMA on the tunately, an antibody specific for the detection of p53Ser46 in expression of ATF-2 and ATF-3 in RAW264.7 mouse macrophages mouse cells is currently not available. High amounts of p21 protein were unexpected and contrary to those observed in human HeLa and were detected in the cytoplasmic fraction of RAW264.7 cells tHPC. Therefore, we questioned whether these transcription factors compared with the faint bands present in some nuclear fractions. It could be activated by a stimulus different than camptothecin or seemed as if p21 was upregulated in cells exposed to the higher TEGDMA. In fact, exposure of RAW264.7 cells to the bacterial Author's personal copy

1792 S. Krifka et al. / Biomaterials 32 (2011) 1787e1795

Fig. 5. Expression of transcription factors in LPS-treated RAW264.7 mouse macro- phages. Cell cultures were exposed to lipopolysaccharide (LPS; 0.1 mg/ml), campto- thecin (1 mM), and TEGDMA (1 and 3 mM) for 24 h, and proteins in cell nuclei were detected as described in Materials and Methods. Medium ¼ untreated cell cultures; control ¼ HeLa cell nuclei isolated from cultures exposed to camptothecin (1 mM) for 24 h. The expression of Lamin A/C was used as a marker of cell nuclei. Typical results from one of two independent experiments are shown.

Fig. 4. Expression of transcription factors in RAW264.7 mouse macrophages. Cell cultures were exposed to 1 and 3 mM TEGDMA and 1 mM camptothecin for 24 h, and the exposure period. Long exposure of RAW264.7 cell cultures to proteins in cell nuclei and the cytoplasmic cell fraction were detected as described in TEGDMA resulted in a substantial increase in the phosphorylation of Materials and Methods. Medium ¼ untreated cell cultures. (A) ¼ transcription factors related to cell responses after DNA damage; (B) ¼ transcription factors related to cell both ERK1/2 and p38 kinases. It was assumed that this activation responses through mitogen-activated protein kinases (MAPK). Phosphorylated was related to the induction of apoptosis as observed after the same proteins were detected by immunoblotting using specific antibodies, and the expres- exposure period [17]. Cell death via apoptosis was detected in the sion of the total amount of proteins (phosphorylated and non-phosphorylated) was presence of TEGDMA in mouse macrophages and human pulp- also monitored on the same membranes after stripping and reprobing with specific derived cells similar to the observations in the present investigation. antibodies. The expression of Lamin A/C and GAPDH was used as a marker of cell nuclei and the cytoplasmic cell fraction. Typical results from one of at least two independent Here, increasing concentrations of TEGDMA reduced the number of experiments are shown. viable mouse macrophages, and the percentage of cells in early and late apoptosis increased in parallel. TEGDMA was less effective in transformed human pulp-derived cells (tHPC), and HeLa were most endotoxin lipopolysaccharide (LPS) strongly increased the expres- resistant to TEGDMA and camptothecin as well. It is suggested that sion of ATF-3 in cell nuclei compared to untreated cell cultures, and the greater susceptibility of RAW264.7 macrophages to TEGDMA the expression of ATF-2 was activated as well (Fig. 5). A decrease in might be related to the downregulation of stress-activated tran- ATF-2 and ATF-3 expression in cultures exposed to camptothecin and scription factors like ATF-2 and ATF-3 as discussed below. TEGDMA in the same experiment was again observed, while camp- The differences between the effectiveness of TEGDMA and tothecin activated ATF-2 and ATF-3 in HeLa control cultures (Fig. 5). camptothecin in RAW264.7 macrophages compared to tHPC and HeLa indicate different mechanisms in the activity of both 4. Discussion substances depending on the cell line. It was reported that camp- tothecin acts during DNA synthesis via strand scission, thus causing 4.1. TEGDMA-induced apoptosis in mouse and human cell lines cell death via apoptosis during the S-phase of the cell cycle [31].In contrast, the current hypothesis for the induction of cell death by Dental resin monomers like TEGDMA and HEMA released from resin monomers like TEGDMA favors an indirect mechanism dental composite materials elicit a wide variety of responses in through the generation of ROS [3]. The observations of TEGDMA- eukaryotic cells in vitro including cytotoxicity through apoptosis, induced apoptosis in the various cell lines in the present investi- genotoxicity, inhibition of the mineralization process and distur- gation correspond with results obtained from primary human bance of functions of the innate immune system. We adhere to the pulp-derived cells and gingival fibroblasts after long exposure hypothesis that these diverse effects on vital cell responses are [17,32,33]. Moreover, the concentration-dependent decrease in the triggered at least in part by the established production of reactive number of vital cells after long exposure periods precisely corre- oxygen species (ROS) indicating the interference of resin monomers lated with the activation of MAPK in the presence of TEGDMA in with a complex redox-sensitive regulation of specific cell functions both mouse macrophages and human pulp-derived cells [17]. [3]. However, thus far, only the role of mitogen-activated protein It has been previously suggested that TEGDMA-induced kinases in the induction of apoptosis has been studied in some detail apoptosis in primary pulp cells might not be mediated through [6,15e17]. It was recently found that extracellular signal-regulated ERK1/2 activation since the percentage of cells in apoptosis only kinases (ERK1/2) and the stress-activated kinases p38 and JNK were slightly changed after pharmacological inhibition of this pathway differentially phosphorylated in mouse macrophages depending on [33]. In addition, there is some evidence that ERK1/2 might be Author's personal copy

S. Krifka et al. / Biomaterials 32 (2011) 1787e1795 1793 activated as a pro-survival factor in cells exposed to high TEGDMA possibly a slight increase in tHPC exposed to TEGDMA. The func- concentrations, and sustained activation of the stress kinases p38 tional consequence of phosphorylation of Ser46 has been reported and perhaps JNK is related to TEGDMA-induced apoptosis in to be related to the onset of p53edependent apoptosis after DNA various cell lines [17,33]. We observed in a recent study that ERK1/2 damage through the activation of apoptotic target genes like PUMA and p38 were activated in THP-1 cells after exposure to TEGDMA for and Noxa [37]. This pathway might also be the molecular basis 24 h and 48 h. This activation was reduced in the presence of the behind the effects of monomers like TEGDMA and is under current ROS scavenger NAC [6]. Thus, it appears as if activation of MAPK by investigation. Here, the expression of p53Ser15 was clearly detec- TEGDMA is possible through the generation of oxidative stress in ted in HeLa cells, and present in tHPC and RAW264.7 macrophages cells of the innate immune system. indicating phosphorylation of p53 by ATM as a response to DNA Here we hypothesized that, in case differential activation of cell damage [18]. Recently, the expression of ATM and p53Ser15 has responses to monomer-induced stress was mediated through also been reported in a salivary gland cell line after long exposure to different MAPKs, appropriate downstream target genes would be HEMA [38]. Besides the p53-independent regulation of p21 activated as well. Therefore, the monomer-induced expression and through, for instance, Sp1 or transforming growth factor b (TGF-b), phosphorylation of transcription factors associated with the phosphorylation of p53 at Ser15 upon DNA damage is discussed as induction of apoptosis was investigated in the present study to a modification which activates the transcription of the cyclin- broaden our understanding of the molecular mechanisms of dental dependent kinase inhibitor 1 (p21) gene [18]. The p21 protein was resin monomer-induced cell death in various cell lines. In addition, detected here in the cytoplasmic as well as the nuclear fraction of monomer-induced apoptosis may originate from DNA damage and untreated cells. In most fractions, the expression of p21 was a subsequent activation of a signaling cascade including p53 and enhanced in nuclei and the cytoplasm of cells treated with camp- p21 [18]. The favorite model of activation of the tumor suppressor tothecin or TEGDMA. The p21 protein plays a dual role in cell cycle protein p53, a transcription factor activated upon DNA damage, progression by regulating both DNA synthesis and CDK activity, but involves signaling via ATM/ATR and checkpoint kinases ChK1/ p21 has also been discussed as an enhancer of cell survival. For ChK2. As the guardian of the genome, p53 is a key regulatory example, it was reported that differentiated cells like macrophages protein of the cell cycle, DNA repair, and the induction of apoptosis contain high amounts of p21, and low levels were associated with due to DNA damage [34,35]. While p53 activity is regulated apoptosis. The cellular localization of p21 is part of the control of its upstream by external stress factors, both at the transcriptional and various functions. While nuclear p21 functions are related to cell posttranslational level, it in turn regulates a wide variety of factors cycle progression, the cytoplasmic p21 pool may be involved in the downstream. Phosphorylation of p53 at Ser15 is linked to the maintenance of cell survival. p21 probably forms a complex with induction of the expression of the cyclin-dependent kinase (CDK) the apoptosis signal-regulating kinase 1 (ASK1) in the cytoplasm to inhibitor p21 which inhibits cell proliferation. Phosphorylation at inhibit stress-related activation of MAPK. It has been shown that Ser46, on the other hand, is thought to be related to the regulation p21 may protect cells from death after exposure to cytotoxic agents of p53-dependent apoptosis [18,36]. and ionizing radiation [39,40]. Thus, it is possible that the relative high levels of p21 detected in the cytoplasmic fraction in the 4.2. TEGDMA-induced activation of transcription factors related to present study act as a defense mechanism against the cell damage DNA damage caused by TEGDMA. This finding might be associated with the detection of the cleavage product of lamin A in RAW264.7 cell While the expression of the p53 protein was greatly increased in cultures exposed to TEGDMA or HeLa exposed to camptothecin. the nuclear fraction of HeLa cells after exposure to camptothecin, Lamin A/C and GAPDH were used as markers to identify the the resin monomer TEGDMA enhanced p53 expression only slightly expression of proteins analyzed here in the nuclear and cyto- at a high concentration. Interestingly, p53 was not detected in the plasmic cell fractions. Lamin cleavage, however, as detected here on cytoplasmic fraction of exposed HeLa cells. Camptothecin also some occasions, is also considered to play a role in the nuclear enhanced the expression of p53 in cell nuclei of tHPC and apoptotic process [41]. RAW264.7 macrophages, and TEGDMA was slightly effective. Camptothecin also stimulated the formation of p53 phosphoryla- 4.3. TEGDMA-induced activation of transcription factors related to tion at Ser15 and Ser46 in cell nuclei, while this activation of p53 the activation of MAPK was apparently very weak in TEGDMA-treated HeLa cells under the current experimental conditions. These results indicate that p53 Cellular stress response is mediated through the activation of activation by the resin monomer is probably lower than the effect signal transduction pathways that may also modify the expression of camptothecin, although TEGDMA was active in treated cultures of immediate early genes. These genes encode transcription factors of the various cell lines as indicated by the reduction in viable cells. which consecutively regulate the activation of downstream genes, Our observations with HeLa and RAW264.7 macrophages are finally leading to apoptosis or supporting cell survival. It has been consistent with the reported low levels of p53 in non-stressed cells reported that the induction of immediate early genes like c-Jun and owing to degradation via proteasomes and the induction of p53 c-Fos, and activating transcription factors ATF-2 and ATF-3 was expression in response to genotoxic agents. It is likely that p53 inducible by a wide variety of external signals including serum, accumulated in the cell nucleus upon DNA damage caused by growth factors, active phorbol esters, ionizing radiation, pharma- camptothecin, and to some extent by TEGDMA [18]. The relative cological agents, stress and cytokines. The MAP kinases ERK1/2, the high expression of p53 in both nuclei and the cytoplasmic fraction stress-activated protein kinases JNK (c-Jun N-terminal kinase) and of tHPC is most likely a consequence of the transformation process p38 may function as second messenger pathways [20]. of the pulp-derived cells. Transformation of the cells with the viral The transcription factor complex AP-1, which is a homo- or oncogene SV40 T-Ag resulted in the extension of the cells’ life span heterodimer composed of Fos/Jun and ATF family members, is due to the binding of SV40 T-Ag to p53 and pRb [29]. Yet, both activated through phosphorylation of Jun by JNK. The JNK/Jun camptothecin and TEGDMA increased the expression of p53 in pathway regulates a vast number of target genes containing AP1- tHPC to some extent. binding sites, including genes that control the cell cycle, as well as The present findings indicated a weak increase in the formation survival and apoptosis [20,42]. In the present study, camptothecin of p53 Ser46 in cell nuclei of HeLa exposed to camptothecin, and increased the expression of the proto-oncoprotein c-Jun in HeLa Author's personal copy

1794 S. Krifka et al. / Biomaterials 32 (2011) 1787e1795 cell nuclei while phospho-c-Jun and c-Jun were not noticeably 5. Conclusion increased by the monomer TEGDMA. This observation is in line with a previous report on camptothecin as an inhibitor of DNA The present study provides experimental evidence that topoisomerase that suggested a link between topoisomerase- TEGDMA interferes with the regulation of cellular pathways mediated DNA damage and intra-cellular signaling events through through transcription factors activated as a consequence of DNA a JNK pathway in HeLa [43]. In contrast, the expression of c-Jun in damage, or initiated downstream from MAPK as suggested in linear tHPC was detected in the nucleic and cytoplasmic fraction, and signaling models. Differential activation of these transcription TEGDMA probably inhibited the phosphorylation of c-Jun. Other factors was detected in cells derived from organs as diverse as the investigations demonstrated that phosphorylation of c-Jun through innate immune system or human pulp tissue after a long exposure JNK could stabilize the protein and in that way contribute to its period. The activation of the tumor suppressor protein p53 by transcriptional activity. On the other hand, phosphorylation of TEGDMA in the various cell lines and its negative effect on the c-Jun was also discussed as the signal for degradation through the expression of transcription factors linked to MAPK activity was ubiquitin pathway [20]. We have no plausible hypothesis at present unknown thus far. The time-related induction of apoptosis and the regarding the reduced expression of both c-Jun and phospho-c-Jun modification of the various transcription factors in the presence of in RAW264.7 macrophages exposed to the genotoxic agents TEGDMA is also a novel finding. Yet, the direct causal correlation camptothecin and TEGDMA. However, it was speculated that the between DNA damage, activation or inhibition of MAPKs and major apoptotic pathway in neurons through JNK-c-Jun may be transcription factors, and apoptosis is still under current investi- inhibited by degradation of c-Jun, thereby allowing neurons to gation using pharmacological and genetic inhibitors. However, the tolerate apoptotic levels of JNK activation [20]. Since c-Jun is an present study indicates that induction of apoptosis in different cell ATF-2 target gene, it is also possible that the reduced expression of lines in the presence of monomers like TEGDMA may be subject to c-Jun is the consequence of a downregulation of ATF-2 [24]. a higher level of complexity than currently suggested by simple ATF-2 (activating transcription factor 2) is a member of the basic linear models. region leucine zipper (bZIP) family of proteins and part of activating protein-1 (AP-1) [26]. It is phosphorylated through stress-activated Acknowledgment protein kinases JNK or p38 in the presence of stressors like pro- inflammatory cytokines, UV irradiation, DNA damage or changes in The authors are grateful to Mrs. Christine Ross-Cavanna for the level of reactive oxygen species (ROS). Nonetheless, ATF-2 is a critical reading of the manuscript. This study was supported by also activated via the Ras-ERK pathway. Diverse signaling in the the Deutsche Forschungsgemeinschaft DFG (grant number Schw activation of ATF-2 is thought to result in a diversity of ATF-2 het- 431/13-1). erodimeric partners activated in a stimulus-specific manner. Among other functions, ATF-2 is required for cellular stress and Appendix DNA damage response including the protection of cells from ionizing irradiation [24]. It appears as if ATF-2 expression solely in Figures with essential color discrimination. Fig. 1 in this article is cell nuclei was downregulated to some extent in tHPC after expo- difficult to interpret in black and white. The full color images can be sure to camptothecin and TEGDMA, and clearly in HeLa and found in the on-line version, at doi:10.1016/j.biomaterials.2010.11.031. RAW264.7 macrophages. As a target of JNK/p38 signaling, phos- phorylation of ATF-2 was increased in RAW264.7 macrophages References after stimulation with lipopolysaccharide (LPS). Consequently, it was suggested that ATF-2 might be involved in the transcriptional [1] Seiss M, Langer C, Hickel R, Reichl FX. Quantitative determination of TEGDMA, control mediated by Toll-like receptors (TLRs) in mouse macro- BHT, and DMABEE in eluates from polymerized resin-based dental restorative e phages [44]. We have recently shown that LPS-stimulated cytokine materials by use of GC/MS. Arch Toxicol 2009;83:1109 15. [2] Santerre JP, Shajii L, Leung BW. Relation of dental composite formulations to production was instantaneously downregulated in the presence of their degradation and the release of hydrolyzed polymeric-resin-derived TEGDMA [17]. Since an inhibition of ATF-2 expression was also products. Crit Rev Oral Biol Med 2001;12:136e51. found in the present study, we speculate that ATF-2 might be [3] Schweikl H, Spagnuolo G, Schmalz G. Genetic and cellular toxicology of dental resin monomers. J Dent Res 2006;85:870e7. involved in TEGDMA-induced inhibition of cytokine production in [4] Stanislawski L, Lefeuvre M, Bourd K, Soheili-Majd E, Goldberg M, Perianin A. macrophages. This finding is underscored by the fact that TEGDMA-induced toxicity in human fibroblasts is associated with early and RAW264.7 macrophages responded to LPS by enhanced ATF-2 and drastic glutathione depletion with subsequent production of oxygen reactive species. J Biomed Mater Res A 2003;66:476e82. ATF-3 expression in the present investigation. These observations [5] Spagnuolo G, Mauro C, Leopardi A, Santillo M, Paterno R, Schweikl H, et al. appear to be in line with those reported by other investigators as NF-kappaB protection against apoptosis induced by HEMA. J Dent Res 2004; well [44,45]. 83:837e42. [6] Eckhardt A, Gerstmayr N, Hiller KA, Bolay C, Waha C, Spagnuolo G, et al. ATF-3 is another member of the ATF/CREB family of transcrip- TEGDMA-induced oxidative DNA damage and activation of ATM and MAP tion factors among the immediate early genes without any obvious kinases. Biomaterials 2009;30:2006e14. tissue-specific or stimulus-specific activation. However, it has been [7] Chang HH, Guo MK, Kasten FH, Chang MC, Huang GF, Wang YL, et al. Stim- ‘ ’ ulation of glutathione depletion, ROS production and cell cycle arrest of dental suggested that ATF-3 may be an adaptive response gene respon- pulp cells and gingival epithelial cells by HEMA. Biomaterials 2005; sible for the coordination of cell reactions to extra- and/or intra- 26:745e53. cellular changes. To date, the biological role of ATF-3 remains [8] Schweikl H, Hartmann A, Hiller K-A, Spagnuolo G, Bolay C, Brockhoff G, et al. obscure, although it has been reported to be involved with the cell Inhibition of TEGDMA and HEMA-induced genotoxicity and cell cycle arrest by N-acetylcysteine. Dent Mater 2007;23:688e95. cycle control, apoptosis and the stress response [46]. The activation [9] Yamada M, Kojima N, Paranjpe A, Att W, Aita H, Jewett A, et al. N-acetyl of ATF-3 in HeLa and tHPC in the nuclear fraction of camptothecin- cysteine (NAC)-assisted detoxification of PMMA resin. J Dent Res 2008; e treated HeLa and tHPC cultures suggests that ATF-3 plays a func- 87:372 7. [10] Galler K, Schweikl H, Hiller KA, Cavender A, Bolay C, D’Souza R, et al. TEGDMA tional role in the stress response, while the consequences of its reduces the expression of genes involved in biomineralization, J Dent Res, in downregulation in RAW264.7 macrophages remain unclear at press. present. We are currently pursuing the possibility that TEGDMA [11] Yamada M, Ueno T, Minamikawa H, Sato N, Iwasa F, Hori N, et al. N-acetyl cysteine alleviates cytotoxicity of bone substitute. J Dent Res 2010;89:411e6. might interfere with the suggested role of ATF-3 to regulate TLR- [12] Eckhardt A, Harorli T, Limtanyakul J, Hiller KA, Bosl C, Bolay C, et al. Inhibition stimulated inflammatory responses [47]. of cytokine and surface antigen expression in LPS-stimulated murine Author's personal copy

S. Krifka et al. / Biomaterials 32 (2011) 1787e1795 1795

macrophages by triethylene glycol dimethacrylate. Biomaterials 2009; early apoptotic cells using fluorescein labelled Annexin V. J Immunol Methods 30:1665e74. 1995;184:39e51. [13] Mantellini MG, Botero TM, Yaman P, Dennison JB, Hanks CT, Nör JE. Adhesive [31] Legarza K, Yang LX. New molecular mechanisms of action of camptothecin- resin induces apoptosis and cell-cycle arrest of pulp cells. J Dent Res type drugs. Anticancer Res 2006;26(5A):3301e5. 2003;82:592e6. [32] Janke V, von Neuhoff N, Schlegelberger B, Leyhausen G, Geurtsen W. TEGDMA [14] Matsuzawa A, Ichijo H. Stress-responsive protein kinases in redox-regulated causes apoptosis in primary human gingival fibroblasts. J Dent Res apoptosis signaling. Antioxid Redox Signal 2005;7:472e81. 2003;82:814e8. [15] Samuelsen JT, Dahl JE, Karlsson S, Morisbak E, Becher R. Apoptosis induced by [33] Spagnuolo G, Galler K, Schmalz G, Cosentino C, Rengo S, Schweikl H. Inhibition the monomers HEMA and TEGDMA involves formation of ROS and differential of phosphatidylinositol 3-kinase amplifies TEGDMA-induced apoptosis in activation of the MAP-kinases p38, JNK and ERK. Dent Mater 2007;23:34e9. primary human pulp cells. J Dent Res 2004;83:703e7. [16] Spagnuolo G, D’Antò V, Valletta R, Strisciuglio C, Schmalz G, Schweikl H, et al. [34] Nyberg KA, Michelson RJ, Putnam CW, Weinert TA. Toward maintaining the Effect of HEMA on human pulp cells survival pathways ERK and Akt. J Endod genome: DNA damage and replication checkpoints. Annu Rev Genet 2008;34:684e8. 2002;36:617e56. [17] Krifka S, Petzel C, Hiller KA, Frank EM, Bosl C, Spagnuolo G, et al. Resin [35] Roos WP, Kaina B. DNA damage-induced cell death by apoptosis. Trends Mol monomer-induced differential activation of MAP kinases and apoptosis in Med 2006;12:440e50. mouse macrophages and human pulp cells. Biomaterials 2010;31:2964e75. [36] Gartel AL. p21(WAF1/CIP1) and cancer: a shifting paradigm? Biofactors 2009; [18] Lavin MF, Gueven N. The complexity of p53 stabilization and activation. Cell 35:161e4. Death Differ 2006;13:941e50. [37] Yu J, Zhang LPUMA. A potent killer with or without p53. Oncogene 2009; [19] Zhou BP, Liao Y, Xia W, Spohn B, Lee MH, Hung MC. Cytoplasmic localization 27:S71e83. of p21Cip1/WAF1 by Akt-induced phosphorylation in HER-2/neu-over- [38] Samuelsen JT, Holme JA, Becher R, Karlsson S, Morisbak E, Dahl JE. HEMA expressing. Nat Cell Biol 2001;3:245e52. reduces cell proliferation and induces apoptosis in vitro. Dent Mater 2008; [20] Whitmarsh AJ. Regulation of gene transcription by mitogen-activated protein 24:134e40. kinase signaling pathways. Biochim Biophys Acta 2007;1773:1285e98. [39] Asada M, Yamada T, Ichijo H, Delia D, Miyazono K, Fukumuro K, et al. p21Cip1/TI: [21] Brenner DA, O’Hara M, Angel P, Chojkier M, Karin M. Prolonged activation of jun apoptosis inhibitory activity of cytoplasmic p21(Cip1/WAF1) in monocytic and collagenase genes by tumour necrosis factor-alpha. Nature 1989;337:661e3. differentiation. EMBO J 1991;18:1223e34. [22] Devary Y, Gottlieb RA, Lau LF, Karin M. Rapid and preferential activation of the [40] Zhan J, Easton JB, Huang S, Mishra A, Xiao L, Lacy ER, et al. Negative regulation c-jun gene during the mammalian UV response. Mol Cell Biol 1991;11:2804e11. of ASK1 by p21Cip1 involves a small domain that includes Serine 98 that is [23] Derijard B, Hibi M, Wu IH, Barrett T, Su B, Deng T, et al. JNK1: a protein kinase phosphorylated by ASK1 in vivo. Mol Cell Biol 2007;27:3530e41. stimulated by UV light and Ha-Ras that binds and phosphorylates the c-Jun [41] Rao L, Perez D, White E. Lamin proteolysis facilitates nuclear events during activation domain. Cell 1994;76:1025e37. apoptosis. J Cell Biol 1996;135(6 Pt 1):1441e55. [24] Bhoumik A, Pablo Lopez-Bergami P, Ronai Z. ATF2 on the double e activating [42] Wagner EF, Nebreda AR. Signal integration by JNK and p38 MAPK pathways in transcription factor and DNA damage response protein. Pigment Cell Res cancer development. Nat Rev Cancer 2009;9:537e49. 2007;20:498e506. [43] Guo X, Zhang J, Fu X, Wei Q, Lu Y, Li Y, et al. Analysis of common gene expression [25] Maekawa T, Shinagawa T, Sano Y, Sakuma T, Nomura S, Nagasaki K, et al. patterns in four human tumor cell lines exposed to camptothecin using cDNA Reduced levels of ATF-2 predispose mice to mammary tumors. Mol Cell Biol microarray: identification of topoisomerase-mediated DNA damage response 2007;27:1730e44. pathways. Fron Biosci 2006;11:1924e31. [26] Bhoumik A, Ronai Z. ATF2: a transcription factor that elicits oncogenic or [44] Hirose N, Maekawa T, Shinagawa T, Ishii S. ATF-2 regulates lipopolysaccha- tumor suppressor activities. Cell Cycle 2008;7:2341e5. ride-induced transcription in macrophage cells. Biochem Biophys Res Com- [27] Yan C, Lu D, Hai T, Boyd DD. Activating transcription factor 3, a stress sensor, mun 2009;385:72e7. activates p53 by blocking its ubiquitination. EMBO J 2005;24:2425e35. [45] Takii R, Inouye S, Fujimoto M, Nakamura T, Shinkawa T, Prakasam R, et al. [28] Yoshii E. Cytotoxic effects of acrylates and methacrylates: relationships of Heat shock transcription factor 1 inhibits expression of IL-6 through activating monomer structures and cytotoxicity. J Biomed Mater Res 1997;37:517e24. transcription factor 3. J Immunol 2010;184:1041e8. [29] Galler K, Schweikl H, Thonemann B, Schmalz G. Human pulp-derived cells [46] Lu D, Chen J, Hai T. The regulation of ATF3 gene expression by mitogen- immortalized with SV40. Eur J Oral Sci 2006;114:138e46. activated protein kinases. Biochem J 2007;401:559e67. [30] Vermes I, Haanen C, Steffens-Nakken H, Reutelingsperger C. A novel assay for [47] Thompson MR, Xu D, Williams BR. ATF3 transcription factor and its emerging apoptosis. flow cytometric detection of phosphatidylserine expression on roles in immunity and cancer. J Mol Med 2009;87:1053e60.